From: Jason Gunthorpe <jgg@ziepe.ca>
To: Vasant Hegde <vasant.hegde@amd.com>
Cc: iommu@lists.linux.dev, joro@8bytes.org,
suravee.suthikulpanit@amd.com, wei.huang2@amd.com,
jsnitsel@redhat.com
Subject: Re: [PATCH v3 02/12] iommu/amd: Do not override PASID entry in GCR3 table
Date: Tue, 7 Nov 2023 09:34:03 -0400 [thread overview]
Message-ID: <20231107133403.GA4634@ziepe.ca> (raw)
In-Reply-To: <4eb067ed-717b-0649-e119-fe43620d39c2@amd.com>
On Tue, Nov 07, 2023 at 11:56:26AM +0530, Vasant Hegde wrote:
> diff --git a/drivers/iommu/amd/iommu.c b/drivers/iommu/amd/iommu.c
> index 691c040216e3..558e0c196172 100644
> --- a/drivers/iommu/amd/iommu.c
> +++ b/drivers/iommu/amd/iommu.c
> @@ -1879,6 +1879,9 @@ static int __set_gcr3(struct iommu_dev_data *dev_data,
> if (pte == NULL)
> return -ENOMEM;
>
> + if (*pte & GCR3_VALID)
> + return -EBUSY;
> +
> *pte = (gcr3 & PAGE_MASK) | GCR3_VALID;
> amd_iommu_dev_flush_pasid_all(dev_data, pasid);
>
>
>
> I believe as a sane design it should first remove existing mapping (that's why
> we have remove_dev_pasid() interface) before adding new mapping. That's why I
> introduced this patch.
Again, no, all these ops are replace. The ideal driver should simply
write the new GCR3 entry with the new value in a single step. No
detach/attach cycle.
Jason
next prev parent reply other threads:[~2023-11-07 13:34 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-10-16 10:43 [PATCH v3 00/12] iommu/amd: SVA Support (Part 4) - SVA and IOPF Vasant Hegde
2023-10-16 10:43 ` [PATCH v3 01/12] iommu/amd: Rename amd_iommu_v2_supported() as amd_iommu_pasid_supported() Vasant Hegde
2023-11-06 17:50 ` Jason Gunthorpe
2023-10-16 10:43 ` [PATCH v3 02/12] iommu/amd: Do not override PASID entry in GCR3 table Vasant Hegde
2023-11-06 17:51 ` Jason Gunthorpe
2023-11-07 6:26 ` Vasant Hegde
2023-11-07 13:34 ` Jason Gunthorpe [this message]
2023-10-16 10:43 ` [PATCH v3 03/12] iommu/amd: Introduce per device DTE update function Vasant Hegde
2023-11-06 17:54 ` Jason Gunthorpe
2023-11-07 6:47 ` Vasant Hegde
2023-11-07 13:36 ` Jason Gunthorpe
2023-10-16 10:43 ` [PATCH v3 04/12] iommu/amd: Add support for enabling/disabling IOMMU features Vasant Hegde
2023-11-06 17:55 ` Jason Gunthorpe
2023-10-16 10:43 ` [PATCH v3 05/12] iommu/amd: Initial SVA support for AMD IOMMU Vasant Hegde
2023-11-06 23:18 ` Jason Gunthorpe
2023-12-20 11:00 ` Vasant Hegde
2023-10-16 10:43 ` [PATCH v3 06/12] iommu/amd: Add support to enable/disable PASID feature Vasant Hegde
2023-10-16 10:43 ` [PATCH v3 07/12] iommu/amd: Move PPR-related functions into ppr.c Vasant Hegde
2023-10-16 10:43 ` [PATCH v3 08/12] iommu/amd: Define per-IOMMU iopf_queue Vasant Hegde
2023-10-16 10:43 ` [PATCH v3 09/12] iommu/amd: Add support for page response Vasant Hegde
2023-10-16 10:43 ` [PATCH v3 10/12] iommu/amd: Add support for add/remove device for IOPF Vasant Hegde
2023-10-16 10:43 ` [PATCH v3 11/12] iommu/amd: Add IO page fault notifier handler Vasant Hegde
2023-11-06 23:20 ` Jason Gunthorpe
2023-11-07 6:42 ` Vasant Hegde
2023-11-07 13:35 ` Jason Gunthorpe
2023-10-16 10:43 ` [PATCH v3 12/12] iommu/amd: Introduce logic to enable/disable IOPF Vasant Hegde
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20231107133403.GA4634@ziepe.ca \
--to=jgg@ziepe.ca \
--cc=iommu@lists.linux.dev \
--cc=joro@8bytes.org \
--cc=jsnitsel@redhat.com \
--cc=suravee.suthikulpanit@amd.com \
--cc=vasant.hegde@amd.com \
--cc=wei.huang2@amd.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox